NLP_text_classification_v4.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyODLGxuxZ+jEIXswJt1KyiY",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/claudio1975/SWISSTEXT_2022/blob/main/NLP_text_classification_v4.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **NLP_&_INSURANCE: TEXT_CLASSIFICATION**"
      ],
      "metadata": {
        "id": "liWb_eKRH-sQ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![](https://www.babel.sg/wp-content/uploads/2019/11/Insurance-Recommendation-System-Increased-Customer-Conversion-Engagement-Rate.png)"
      ],
      "metadata": {
        "id": "iDPfXei3_LAe"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Import Libraries"
      ],
      "metadata": {
        "id": "mP5_Yn4wCaru"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install ktrain"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bJvjkBX74054",
        "outputId": "72c3bde3-76ab-41a4-c5e5-fce6e1e39ce7"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting ktrain\n",
            "  Downloading ktrain-0.33.0.tar.gz (25.3 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m25.3/25.3 MB\u001b[0m \u001b[31m27.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.8/dist-packages (from ktrain) (1.0.2)\n",
            "Requirement already satisfied: matplotlib>=3.0.0 in /usr/local/lib/python3.8/dist-packages (from ktrain) (3.2.2)\n",
            "Requirement already satisfied: pandas>=1.0.1 in /usr/local/lib/python3.8/dist-packages (from ktrain) (1.3.5)\n",
            "Requirement already satisfied: fastprogress>=0.1.21 in /usr/local/lib/python3.8/dist-packages (from ktrain) (1.0.3)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.8/dist-packages (from ktrain) (2.25.1)\n",
            "Requirement already satisfied: joblib in /usr/local/lib/python3.8/dist-packages (from ktrain) (1.2.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.8/dist-packages (from ktrain) (21.3)\n",
            "Collecting langdetect\n",
            "  Downloading langdetect-1.0.9.tar.gz (981 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m981.5/981.5 KB\u001b[0m \u001b[31m42.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: jieba in /usr/local/lib/python3.8/dist-packages (from ktrain) (0.42.1)\n",
            "Collecting cchardet\n",
            "  Downloading cchardet-2.1.7-cp38-cp38-manylinux2010_x86_64.whl (265 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m266.0/266.0 KB\u001b[0m \u001b[31m11.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: chardet in /usr/local/lib/python3.8/dist-packages (from ktrain) (4.0.0)\n",
            "Collecting syntok>1.3.3\n",
            "  Downloading syntok-1.4.4-py3-none-any.whl (24 kB)\n",
            "Collecting transformers>=4.17.0\n",
            "  Downloading transformers-4.26.0-py3-none-any.whl (6.3 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m6.3/6.3 MB\u001b[0m \u001b[31m26.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting sentencepiece\n",
            "  Downloading sentencepiece-0.1.97-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.3/1.3 MB\u001b[0m \u001b[31m25.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting keras_bert>=0.86.0\n",
            "  Downloading keras-bert-0.89.0.tar.gz (25 kB)\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting whoosh\n",
            "  Downloading Whoosh-2.7.4-py2.py3-none-any.whl (468 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m468.8/468.8 KB\u001b[0m \u001b[31m19.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: numpy in /usr/local/lib/python3.8/dist-packages (from keras_bert>=0.86.0->ktrain) (1.21.6)\n",
            "Collecting keras-transformer==0.40.0\n",
            "  Downloading keras-transformer-0.40.0.tar.gz (9.7 kB)\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting keras-pos-embd==0.13.0\n",
            "  Downloading keras-pos-embd-0.13.0.tar.gz (5.6 kB)\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting keras-multi-head==0.29.0\n",
            "  Downloading keras-multi-head-0.29.0.tar.gz (13 kB)\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting keras-layer-normalization==0.16.0\n",
            "  Downloading keras-layer-normalization-0.16.0.tar.gz (3.9 kB)\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting keras-position-wise-feed-forward==0.8.0\n",
            "  Downloading keras-position-wise-feed-forward-0.8.0.tar.gz (4.1 kB)\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting keras-embed-sim==0.10.0\n",
            "  Downloading keras-embed-sim-0.10.0.tar.gz (3.6 kB)\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting keras-self-attention==0.51.0\n",
            "  Downloading keras-self-attention-0.51.0.tar.gz (11 kB)\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.8/dist-packages (from matplotlib>=3.0.0->ktrain) (2.8.2)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.8/dist-packages (from matplotlib>=3.0.0->ktrain) (3.0.9)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.8/dist-packages (from matplotlib>=3.0.0->ktrain) (0.11.0)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.8/dist-packages (from matplotlib>=3.0.0->ktrain) (1.4.4)\n",
            "Requirement already satisfied: pytz>=2017.3 in /usr/local/lib/python3.8/dist-packages (from pandas>=1.0.1->ktrain) (2022.7)\n",
            "Requirement already satisfied: regex>2016 in /usr/local/lib/python3.8/dist-packages (from syntok>1.3.3->ktrain) (2022.6.2)\n",
            "Collecting tokenizers!=0.11.3,<0.14,>=0.11.1\n",
            "  Downloading tokenizers-0.13.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.6 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.6/7.6 MB\u001b[0m \u001b[31m72.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.8/dist-packages (from transformers>=4.17.0->ktrain) (4.64.1)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.8/dist-packages (from transformers>=4.17.0->ktrain) (3.9.0)\n",
            "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.8/dist-packages (from transformers>=4.17.0->ktrain) (6.0)\n",
            "Collecting huggingface-hub<1.0,>=0.11.0\n",
            "  Downloading huggingface_hub-0.12.0-py3-none-any.whl (190 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m190.3/190.3 KB\u001b[0m \u001b[31m21.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: six in /usr/local/lib/python3.8/dist-packages (from langdetect->ktrain) (1.15.0)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.8/dist-packages (from requests->ktrain) (2.10)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.8/dist-packages (from requests->ktrain) (1.24.3)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.8/dist-packages (from requests->ktrain) (2022.12.7)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.8/dist-packages (from scikit-learn->ktrain) (3.1.0)\n",
            "Requirement already satisfied: scipy>=1.1.0 in /usr/local/lib/python3.8/dist-packages (from scikit-learn->ktrain) (1.7.3)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.8/dist-packages (from huggingface-hub<1.0,>=0.11.0->transformers>=4.17.0->ktrain) (4.4.0)\n",
            "Building wheels for collected packages: ktrain, keras_bert, keras-transformer, keras-embed-sim, keras-layer-normalization, keras-multi-head, keras-pos-embd, keras-position-wise-feed-forward, keras-self-attention, langdetect\n",
            "  Building wheel for ktrain (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for ktrain: filename=ktrain-0.33.0-py3-none-any.whl size=25313831 sha256=d659769ea6ca9866a3fb2e5838d89826d2f85c73b7709ca4c8c27e3145529436\n",
            "  Stored in directory: /root/.cache/pip/wheels/56/6a/5d/676e3966821297d89a64e6f3ca996c148a6c8b94ee1993be3a\n",
            "  Building wheel for keras_bert (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for keras_bert: filename=keras_bert-0.89.0-py3-none-any.whl size=33516 sha256=c6541800f5e593188823bf77745bab546f981e34daafd22708cc197ec28bb7b7\n",
            "  Stored in directory: /root/.cache/pip/wheels/a1/34/ed/6bbd71716d7bcea30d75e8bc5aeb94f4cb52636295c8343534\n",
            "  Building wheel for keras-transformer (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for keras-transformer: filename=keras_transformer-0.40.0-py3-none-any.whl size=12304 sha256=3cecae1f2c3ea026eaf88844a23f50022dfc7db5ba2c4cb89e68a5d5b73d65c7\n",
            "  Stored in directory: /root/.cache/pip/wheels/07/cd/a7/a8fa93f7e177eee0101fed63179f7a2fa3b53671ffaad82bfd\n",
            "  Building wheel for keras-embed-sim (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for keras-embed-sim: filename=keras_embed_sim-0.10.0-py3-none-any.whl size=3959 sha256=7d214c10926ba7a485ae13cf31925e8919e1d0cb98239f778e88f4852708fa0a\n",
            "  Stored in directory: /root/.cache/pip/wheels/59/bd/9a/ec6e575aaa50687d7af968bde7ce710b542eeaa9ee7978d4ba\n",
            "  Building wheel for keras-layer-normalization (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for keras-layer-normalization: filename=keras_layer_normalization-0.16.0-py3-none-any.whl size=4668 sha256=a6c1e62d13728c34c4d803dec38b6b0d56c3e1cbad3c5102c3fb693f1e6a4286\n",
            "  Stored in directory: /root/.cache/pip/wheels/d7/2b/f4/28f4bab995fa99c26b761bc7f9aeb5bf6c81e9be6ccd0b853b\n",
            "  Building wheel for keras-multi-head (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for keras-multi-head: filename=keras_multi_head-0.29.0-py3-none-any.whl size=14993 sha256=9bef6b3319518f96ccc39885e9bf11e0f03fed549148ef420005698b6af0c864\n",
            "  Stored in directory: /root/.cache/pip/wheels/91/eb/bc/ce4bb467f5a7db6727f148f70bb0e52a62ef7edd41a19c8bdd\n",
            "  Building wheel for keras-pos-embd (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for keras-pos-embd: filename=keras_pos_embd-0.13.0-py3-none-any.whl size=6961 sha256=cb79d2db8c2728d1c8c679acdcb44c63ed585e27df6610003d787431bb72fb4b\n",
            "  Stored in directory: /root/.cache/pip/wheels/91/c4/ff/7e13e4f102c3b7d73ff075a50fe3266f3ec2de898d5493a8a2\n",
            "  Building wheel for keras-position-wise-feed-forward (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for keras-position-wise-feed-forward: filename=keras_position_wise_feed_forward-0.8.0-py3-none-any.whl size=4983 sha256=3445bbb6f1aba3cb7d94d1eefa02bfa90d746ef348136edac326fd4b54e998d3\n",
            "  Stored in directory: /root/.cache/pip/wheels/f2/15/39/59861ed531ef6c7c75810500eb22c68a425f82dde31d68630a\n",
            "  Building wheel for keras-self-attention (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for keras-self-attention: filename=keras_self_attention-0.51.0-py3-none-any.whl size=18913 sha256=0c9158396e9ab15624d7ad70881ec67763a85cb149602aa7c9536242739dd8a6\n",
            "  Stored in directory: /root/.cache/pip/wheels/ac/13/2d/3de7c76f618a8d162884ac5b726a8c2242ad88afa370f1e62f\n",
            "  Building wheel for langdetect (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for langdetect: filename=langdetect-1.0.9-py3-none-any.whl size=993242 sha256=12e5d70a1c93a980381e6f7f8f42a2d25dbad9b0577df917fb69356c37063de3\n",
            "  Stored in directory: /root/.cache/pip/wheels/13/c7/b0/79f66658626032e78fc1a83103690ef6797d551cb22e56e734\n",
            "Successfully built ktrain keras_bert keras-transformer keras-embed-sim keras-layer-normalization keras-multi-head keras-pos-embd keras-position-wise-feed-forward keras-self-attention langdetect\n",
            "Installing collected packages: whoosh, tokenizers, sentencepiece, cchardet, syntok, langdetect, keras-self-attention, keras-position-wise-feed-forward, keras-pos-embd, keras-layer-normalization, keras-embed-sim, keras-multi-head, huggingface-hub, transformers, keras-transformer, keras_bert, ktrain\n",
            "Successfully installed cchardet-2.1.7 huggingface-hub-0.12.0 keras-embed-sim-0.10.0 keras-layer-normalization-0.16.0 keras-multi-head-0.29.0 keras-pos-embd-0.13.0 keras-position-wise-feed-forward-0.8.0 keras-self-attention-0.51.0 keras-transformer-0.40.0 keras_bert-0.89.0 ktrain-0.33.0 langdetect-1.0.9 sentencepiece-0.1.97 syntok-1.4.4 tokenizers-0.13.2 transformers-4.26.0 whoosh-2.7.4\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#setup deterministic results \n",
        "import numpy as np\n",
        "from numpy.random import seed\n",
        "seed=0\n",
        "np.random.seed(seed)\n",
        "import tensorflow as tf\n",
        "tf.random.set_seed(seed)\n",
        "import os\n",
        "os.environ['PYTHONHASHSEED'] = str(seed)\n",
        "os.environ['TF_DETERMINISTIC_OPS'] = '1'\n",
        "\n",
        "import pandas as pd\n",
        "import sklearn\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import f1_score, accuracy_score\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n",
        "from sklearn.model_selection import StratifiedKFold\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.decomposition import PCA\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import wordcloud\n",
        "from wordcloud import WordCloud, STOPWORDS\n",
        "import re\n",
        "import string\n",
        "#import langdetect \n",
        "import ktrain\n",
        "from ktrain import text\n",
        "import spacy\n",
        "import gensim\n",
        "import gensim.corpora as corpora\n",
        "from gensim.models.coherencemodel import CoherenceModel\n",
        "from gensim.models import Word2Vec \n",
        "import gensim.downloader as gensim_api\n",
        "#from bertopic import BERTopic"
      ],
      "metadata": {
        "id": "fs11-otnJH6u"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Requirements"
      ],
      "metadata": {
        "id": "cksn-dSFNzmX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(np.__version__) "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eGmRQbDsJqI-",
        "outputId": "77d2d630-d091-4399-cbba-8eec6ce95db2"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1.21.6\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(pd.__version__) "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ge7646ZbLSkn",
        "outputId": "ff42016f-bf16-4eaf-9ac9-d9e05e043816"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1.3.5\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(sklearn.__version__)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wY4eVqHRLSuK",
        "outputId": "a43608c4-17bc-4240-d543-a314f7a5a9df"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1.0.2\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(sns.__version__)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CQFoDoZJLSw-",
        "outputId": "9d532de6-d277-4c41-8c91-2caa2f381054"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.11.2\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(tf.__version__)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SHkBAE00LS4-",
        "outputId": "e9c47a16-5417-41aa-c0c0-4886aa588c82"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.9.2\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(spacy.__version__)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Nt7-AB_JBeuG",
        "outputId": "fd23022b-ae66-4f4e-8316-730a15d577ca"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "3.4.4\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(gensim.__version__)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "siNtLDbkuVGG",
        "outputId": "1ed9124c-f239-4d91-d14a-84b1a2acba72"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "3.6.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(ktrain.__version__)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O6NB1WKWC1Wc",
        "outputId": "bdce4e9e-eca8-4c5d-e33b-a0a457f5fe69"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.33.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Upload dataset"
      ],
      "metadata": {
        "id": "7cl-fs32LpJ3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# https://www.kaggle.com/competitions/nlp-getting-started"
      ],
      "metadata": {
        "id": "z3LcLDYbz88e"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_train = pd.read_csv('https://raw.githubusercontent.com/claudio1975/SWISSTEXT_2022/main/data/df_train.csv')"
      ],
      "metadata": {
        "id": "TmqNgXQiBiDT"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#from google.colab import files \n",
        "#uploaded = files.upload() "
      ],
      "metadata": {
        "id": "t9QumHJqLS7q"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Reading File\n",
        "#df_train = pd.read_csv('df_train.csv')"
      ],
      "metadata": {
        "id": "zm6ikNARLS-C"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# select columns\n",
        "df_4 = df_train.iloc[:,1:8]\n",
        "df_4.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "rqdmz3Jwvr3s",
        "outputId": "8ea2264f-da03-4a24-e44f-00d7fc95b817"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   id keyword location                                               text  \\\n",
              "0   1     NaN      NaN  Our Deeds are the Reason of this #earthquake M...   \n",
              "1   4     NaN      NaN             Forest fire near La Ronge Sask. Canada   \n",
              "2   5     NaN      NaN  All residents asked to 'shelter in place' are ...   \n",
              "3   6     NaN      NaN  13,000 people receive #wildfires evacuation or...   \n",
              "4   7     NaN      NaN  Just got sent this photo from Ruby #Alaska as ...   \n",
              "\n",
              "   target lang                                       text_cleaned  \n",
              "0       1   en              deeds reason earthquake allah forgive  \n",
              "1       1   en              forest fire near la ronge sask canada  \n",
              "2       1   en  resident ask shelter place notify officer evac...  \n",
              "3       1   en  people receive wildfire evacuation order calif...  \n",
              "4       1   en  get send photo ruby alaska smoke wildfires pou...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e6d0f50b-4784-42c2-b0b9-1d8d1516031c\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>keyword</th>\n",
              "      <th>location</th>\n",
              "      <th>text</th>\n",
              "      <th>target</th>\n",
              "      <th>lang</th>\n",
              "      <th>text_cleaned</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Our Deeds are the Reason of this #earthquake M...</td>\n",
              "      <td>1</td>\n",
              "      <td>en</td>\n",
              "      <td>deeds reason earthquake allah forgive</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Forest fire near La Ronge Sask. Canada</td>\n",
              "      <td>1</td>\n",
              "      <td>en</td>\n",
              "      <td>forest fire near la ronge sask canada</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>All residents asked to 'shelter in place' are ...</td>\n",
              "      <td>1</td>\n",
              "      <td>en</td>\n",
              "      <td>resident ask shelter place notify officer evac...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>6</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>13,000 people receive #wildfires evacuation or...</td>\n",
              "      <td>1</td>\n",
              "      <td>en</td>\n",
              "      <td>people receive wildfire evacuation order calif...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>7</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Just got sent this photo from Ruby #Alaska as ...</td>\n",
              "      <td>1</td>\n",
              "      <td>en</td>\n",
              "      <td>get send photo ruby alaska smoke wildfires pou...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e6d0f50b-4784-42c2-b0b9-1d8d1516031c')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-e6d0f50b-4784-42c2-b0b9-1d8d1516031c button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-e6d0f50b-4784-42c2-b0b9-1d8d1516031c');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Prepare Data sets"
      ],
      "metadata": {
        "id": "v95PdPJAX1I8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_docs = df_4.copy()"
      ],
      "metadata": {
        "id": "K5Fxq2kri4Fm"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_docs = train_docs[['text','target']]\n",
        "train_docs.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "BAxVZUAiekng",
        "outputId": "cd2dd748-1bb1-4726-99cc-87634784bac2"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                text  target\n",
              "0  Our Deeds are the Reason of this #earthquake M...       1\n",
              "1             Forest fire near La Ronge Sask. Canada       1\n",
              "2  All residents asked to 'shelter in place' are ...       1\n",
              "3  13,000 people receive #wildfires evacuation or...       1\n",
              "4  Just got sent this photo from Ruby #Alaska as ...       1"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a9d1ccf7-9e24-4111-8f39-159f76c1bf39\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Our Deeds are the Reason of this #earthquake M...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Forest fire near La Ronge Sask. Canada</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>All residents asked to 'shelter in place' are ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>13,000 people receive #wildfires evacuation or...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Just got sent this photo from Ruby #Alaska as ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a9d1ccf7-9e24-4111-8f39-159f76c1bf39')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-a9d1ccf7-9e24-4111-8f39-159f76c1bf39 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-a9d1ccf7-9e24-4111-8f39-159f76c1bf39');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "classification = {1:'disaster',0:'not_disaster'}\n",
        "train_docs['classification'] = train_docs['target'].map(classification)\n",
        "train_docs_ = train_docs[['text','classification']]\n",
        "# split target variable\n",
        "y = train_docs_['classification']\n",
        "train_docs_.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "m1cp4NXsX4-u",
        "outputId": "1a0b6f64-1c03-459b-cc47-04f4b905db43"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                text classification\n",
              "0  Our Deeds are the Reason of this #earthquake M...       disaster\n",
              "1             Forest fire near La Ronge Sask. Canada       disaster\n",
              "2  All residents asked to 'shelter in place' are ...       disaster\n",
              "3  13,000 people receive #wildfires evacuation or...       disaster\n",
              "4  Just got sent this photo from Ruby #Alaska as ...       disaster"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-d1cbdf64-7a5b-4dd0-8695-decafb7612dd\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>classification</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Our Deeds are the Reason of this #earthquake M...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Forest fire near La Ronge Sask. Canada</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>All residents asked to 'shelter in place' are ...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>13,000 people receive #wildfires evacuation or...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Just got sent this photo from Ruby #Alaska as ...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d1cbdf64-7a5b-4dd0-8695-decafb7612dd')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-d1cbdf64-7a5b-4dd0-8695-decafb7612dd button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-d1cbdf64-7a5b-4dd0-8695-decafb7612dd');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def tt_split(X, y, test_size=0.2):\n",
        "    i = int((1 - test_size) * X.shape[0]) \n",
        "    o = np.random.permutation(X.shape[0])\n",
        "    \n",
        "    X_train, X_test = np.split(np.take(X,o,axis=0), [i])\n",
        "    y_train, y_test = np.split(np.take(y,o), [i])\n",
        "    return X_train, X_test, y_train, y_test"
      ],
      "metadata": {
        "id": "skq4COdadns9"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train_, X_test, y_train_, y_test = tt_split(train_docs_,y)"
      ],
      "metadata": {
        "id": "A7c_WSK4gnpP"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train_"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "qVoACyVCiVwz",
        "outputId": "d44e8685-2e8d-4464-f8b5-bbed4dd8084a"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                   text classification\n",
              "311   @KatieKatCubs you already know how this shit g...   not_disaster\n",
              "4970  @LeMaireLee @danharmon People Near Meltdown Co...   not_disaster\n",
              "527   1-6 TIX Calgary Flames vs COL Avalanche Presea...   not_disaster\n",
              "6362  If you ever think you running out of choices i...   not_disaster\n",
              "800   If you dotish to blight your car go right ahea...   not_disaster\n",
              "...                                                 ...            ...\n",
              "6647  @SwiftyCommissh @TaoistInsight @ImmortalTech S...       disaster\n",
              "6031  Thanks Benson &amp; Clegg for the #follow! Che...   not_disaster\n",
              "239   ANNIHILATED IN DAMASCUS: SYRIAN ARMY GRINDS Û...       disaster\n",
              "5720  VIDEO: 'We're picking up bodies from water': R...       disaster\n",
              "7296  My brother-n-law riooooos got the call to head...       disaster\n",
              "\n",
              "[6090 rows x 2 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1398bb0e-d721-43ef-8f02-e1692762132b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>classification</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>311</th>\n",
              "      <td>@KatieKatCubs you already know how this shit g...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4970</th>\n",
              "      <td>@LeMaireLee @danharmon People Near Meltdown Co...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>527</th>\n",
              "      <td>1-6 TIX Calgary Flames vs COL Avalanche Presea...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6362</th>\n",
              "      <td>If you ever think you running out of choices i...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>800</th>\n",
              "      <td>If you dotish to blight your car go right ahea...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6647</th>\n",
              "      <td>@SwiftyCommissh @TaoistInsight @ImmortalTech S...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6031</th>\n",
              "      <td>Thanks Benson &amp;amp; Clegg for the #follow! Che...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>239</th>\n",
              "      <td>ANNIHILATED IN DAMASCUS: SYRIAN ARMY GRINDS Û...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5720</th>\n",
              "      <td>VIDEO: 'We're picking up bodies from water': R...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7296</th>\n",
              "      <td>My brother-n-law riooooos got the call to head...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>6090 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1398bb0e-d721-43ef-8f02-e1692762132b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1398bb0e-d721-43ef-8f02-e1692762132b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1398bb0e-d721-43ef-8f02-e1692762132b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "z6gOnftZakbH",
        "outputId": "9861e925-d4c7-42a9-85d1-3a9aaff53b9e"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                   text classification\n",
              "542   The possible new jerseys for the Avalanche nex...   not_disaster\n",
              "2725  ÛÏRichmond Coaches were devastated to hear of...       disaster\n",
              "2454  Madhya Pradesh Train Derailment: Village Youth...       disaster\n",
              "1459  Small casualty on the way to Colorado http://t...       disaster\n",
              "5764  'Without an ally near you can't use this skill...   not_disaster\n",
              "...                                                 ...            ...\n",
              "4931  They are the real heroes... RIP Brave hearts.....   not_disaster\n",
              "3264  Car engulfed in flames backs up traffic at Par...       disaster\n",
              "1653  Great British Bake Off's back and Dorret's cho...       disaster\n",
              "2607  Black Eye 9: A space battle occurred at Star O...   not_disaster\n",
              "2732  ???????????? @MikeParrActor absolutely devasta...   not_disaster\n",
              "\n",
              "[1523 rows x 2 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-ecf191d4-588e-4c25-a3a9-73c7fa225ac9\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>classification</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>542</th>\n",
              "      <td>The possible new jerseys for the Avalanche nex...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2725</th>\n",
              "      <td>ÛÏRichmond Coaches were devastated to hear of...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2454</th>\n",
              "      <td>Madhya Pradesh Train Derailment: Village Youth...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1459</th>\n",
              "      <td>Small casualty on the way to Colorado http://t...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5764</th>\n",
              "      <td>'Without an ally near you can't use this skill...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4931</th>\n",
              "      <td>They are the real heroes... RIP Brave hearts.....</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3264</th>\n",
              "      <td>Car engulfed in flames backs up traffic at Par...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1653</th>\n",
              "      <td>Great British Bake Off's back and Dorret's cho...</td>\n",
              "      <td>disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2607</th>\n",
              "      <td>Black Eye 9: A space battle occurred at Star O...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2732</th>\n",
              "      <td>???????????? @MikeParrActor absolutely devasta...</td>\n",
              "      <td>not_disaster</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1523 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ecf191d4-588e-4c25-a3a9-73c7fa225ac9')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-ecf191d4-588e-4c25-a3a9-73c7fa225ac9 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-ecf191d4-588e-4c25-a3a9-73c7fa225ac9');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "(x_train, y_train), (x_val, y_val), preproc = text.texts_from_df(train_df = X_train_, \n",
        "                                                                   text_column = 'text',\n",
        "                                                                   label_columns=['classification'],\n",
        "                                                                   maxlen=100, \n",
        "                                                                   max_features=100000,\n",
        "                                                                   preprocess_mode='bert',\n",
        "                                                                   val_pct=0.2)\n"
      ],
      "metadata": {
        "id": "e42fL9RDNKbt",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 520
        },
        "outputId": "350855b0-a11e-470e-f4f3-acabc0d70b96"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['disaster', 'not_disaster']\n",
            "      disaster  not_disaster\n",
            "6374       1.0           0.0\n",
            "6503       0.0           1.0\n",
            "1988       0.0           1.0\n",
            "4173       0.0           1.0\n",
            "1699       0.0           1.0\n",
            "['disaster', 'not_disaster']\n",
            "      disaster  not_disaster\n",
            "2844       1.0           0.0\n",
            "6691       1.0           0.0\n",
            "1746       0.0           1.0\n",
            "6134       0.0           1.0\n",
            "5757       0.0           1.0\n",
            "downloading pretrained BERT model (uncased_L-12_H-768_A-12.zip)...\n",
            "[██████████████████████████████████████████████████]\n",
            "extracting pretrained BERT model...\n",
            "done.\n",
            "\n",
            "cleanup downloaded zip...\n",
            "done.\n",
            "\n",
            "preprocessing train...\n",
            "language: en\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "done."
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Is Multi-Label? False\n",
            "preprocessing test...\n",
            "language: en\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "<style>\n",
              "    /* Turns off some styling */\n",
              "    progress {\n",
              "        /* gets rid of default border in Firefox and Opera. */\n",
              "        border: none;\n",
              "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "        background-size: auto;\n",
              "    }\n",
              "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
              "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
              "    }\n",
              "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "        background: #F44336;\n",
              "    }\n",
              "</style>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "done."
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Modelling with BERT\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "tGW5DZBPD7Sl"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_BERT = text.text_classifier(name='bert', train_data = (x_train, y_train) , preproc=preproc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "srZyAxKGMj2l",
        "outputId": "fbe43642-c664-4197-e39a-3dad3c48173f"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Is Multi-Label? False\n",
            "maxlen is 100\n",
            "done.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "learner_BERT = ktrain.get_learner(model = model_BERT, \n",
        "                             train_data=(x_train, y_train), \n",
        "                             val_data=(x_val, y_val), \n",
        "                             batch_size=32, \n",
        "                             use_multiprocessing = True)\n"
      ],
      "metadata": {
        "id": "3hBCDBT4ywjb"
      },
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "learner_BERT.fit_onecycle(lr=2e-5, epochs=1,checkpoint_folder='output')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wmGW_oZHzB1a",
        "outputId": "e9a7e986-b61c-4491-b5ae-7092013846e9"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "begin training using onecycle policy with max lr of 2e-05...\n",
            "153/153 [==============================] - 139s 770ms/step - loss: 0.5086 - accuracy: 0.7562 - val_loss: 0.3948 - val_accuracy: 0.8415\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fb7f018d820>"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_BERT = ktrain.get_predictor(learner_BERT.model, preproc)"
      ],
      "metadata": {
        "id": "oUr24X4fYCN3"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "learner_BERT.print_layers()"
      ],
      "metadata": {
        "id": "7PJL3iHxBOXa",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e64fc269-2e85-4c10-949b-8820a53c1f83"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0 (trainable=True) : <keras.engine.input_layer.InputLayer object at 0x7fb80a3767c0>\n",
            "1 (trainable=True) : <keras.engine.input_layer.InputLayer object at 0x7fb80a38cd30>\n",
            "2 (trainable=True) : <keras_bert.layers.embedding.TokenEmbedding object at 0x7fb80a376790>\n",
            "3 (trainable=True) : <keras.layers.core.embedding.Embedding object at 0x7fb7f481f0a0>\n",
            "4 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f4b0fca0>\n",
            "5 (trainable=True) : <keras_pos_embd.pos_embd.PositionEmbedding object at 0x7fb7f07a7820>\n",
            "6 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f0787910>\n",
            "7 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f0787c40>\n",
            "8 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f06b6df0>\n",
            "9 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f06b67f0>\n",
            "10 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f064c1c0>\n",
            "11 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f0748370>\n",
            "12 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f0755880>\n",
            "13 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb8e5c8bb80>\n",
            "14 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f064c730>\n",
            "15 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f06a05e0>\n",
            "16 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f06a08e0>\n",
            "17 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f06a0370>\n",
            "18 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f054e880>\n",
            "19 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f04e40a0>\n",
            "20 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f06c51c0>\n",
            "21 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f06a0c10>\n",
            "22 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f05f9970>\n",
            "23 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f04f6d00>\n",
            "24 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f0642bb0>\n",
            "25 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f0642c40>\n",
            "26 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f06e2880>\n",
            "27 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f0524730>\n",
            "28 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f06b19a0>\n",
            "29 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f0642970>\n",
            "30 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f04f0700>\n",
            "31 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f04d0250>\n",
            "32 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f0577cd0>\n",
            "33 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f04f6610>\n",
            "34 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f04f0130>\n",
            "35 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f06c5b20>\n",
            "36 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f06ded90>\n",
            "37 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f04d02e0>\n",
            "38 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f047f3a0>\n",
            "39 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f06ed040>\n",
            "40 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f074b670>\n",
            "41 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f074bdf0>\n",
            "42 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f04dcd30>\n",
            "43 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f056cf40>\n",
            "44 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f074bb50>\n",
            "45 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f074b4c0>\n",
            "46 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f06edfa0>\n",
            "47 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f047f160>\n",
            "48 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f04054f0>\n",
            "49 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f04056d0>\n",
            "50 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f0408bb0>\n",
            "51 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f042d850>\n",
            "52 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f0313610>\n",
            "53 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f0405c40>\n",
            "54 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f0313430>\n",
            "55 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f031ddc0>\n",
            "56 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f02bc970>\n",
            "57 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f02bc580>\n",
            "58 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f032f850>\n",
            "59 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f02cc430>\n",
            "60 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f027a460>\n",
            "61 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f02bc280>\n",
            "62 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f054f460>\n",
            "63 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f028e730>\n",
            "64 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f02a3580>\n",
            "65 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f02a37f0>\n",
            "66 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f02ad040>\n",
            "67 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f02b27f0>\n",
            "68 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f0253eb0>\n",
            "69 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f0297d60>\n",
            "70 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f4b0f760>\n",
            "71 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f022dfd0>\n",
            "72 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f01c9ca0>\n",
            "73 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f01c9a90>\n",
            "74 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f01c94c0>\n",
            "75 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f02a39d0>\n",
            "76 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f01c9a60>\n",
            "77 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f01c9fa0>\n",
            "78 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f0236eb0>\n",
            "79 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f018dcd0>\n",
            "80 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f01b6520>\n",
            "81 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f01b64f0>\n",
            "82 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f01b6c10>\n",
            "83 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f019aac0>\n",
            "84 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f0161f40>\n",
            "85 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f02977f0>\n",
            "86 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f012e9d0>\n",
            "87 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f01a7be0>\n",
            "88 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f00bf070>\n",
            "89 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f018d970>\n",
            "90 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f018dfd0>\n",
            "91 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f00bf6a0>\n",
            "92 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f018d610>\n",
            "93 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f012e2b0>\n",
            "94 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f00c7580>\n",
            "95 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f00b3a60>\n",
            "96 (trainable=True) : <keras_multi_head.multi_head_attention.MultiHeadAttention object at 0x7fb7f006d7c0>\n",
            "97 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f00b37c0>\n",
            "98 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f00bf160>\n",
            "99 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f005cf40>\n",
            "100 (trainable=True) : <keras_position_wise_feed_forward.feed_forward.FeedForward object at 0x7fb7f01b6880>\n",
            "101 (trainable=True) : <keras.layers.regularization.dropout.Dropout object at 0x7fb7f0196d30>\n",
            "102 (trainable=True) : <keras.layers.merging.add.Add object at 0x7fb7f006d790>\n",
            "103 (trainable=True) : <keras_layer_normalization.layer_normalization.LayerNormalization object at 0x7fb7f01a75b0>\n",
            "104 (trainable=True) : <keras_bert.layers.extract.Extract object at 0x7fb7f0246070>\n",
            "105 (trainable=True) : <keras.layers.core.dense.Dense object at 0x7fb8e5ca6160>\n",
            "106 (trainable=True) : <keras.layers.core.dense.Dense object at 0x7fb7f4b0f9d0>\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pred_tweets_train_BERT = predictor_BERT.predict(X_train_.text.values)\n",
        "pred_tweets_train_BERT = pd.DataFrame(data=pred_tweets_train_BERT, columns=['predicted_label_train'])"
      ],
      "metadata": {
        "id": "_qc9M6ydEPaG",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "77cf632f-ba34-4184-bcb2-351ff494c3ba"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "191/191 [==============================] - 46s 221ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pred_tweets_test_BERT = predictor_BERT.predict(X_test.text.values)\n",
        "pred_tweets_test_BERT = pd.DataFrame(data=pred_tweets_test_BERT, columns=['predicted_label_test'])"
      ],
      "metadata": {
        "id": "yC2ryYdFEbW8",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "503bacf2-04b6-48b8-c090-c3d6689085cd"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "48/48 [==============================] - 10s 218ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "accuracy_tweets_train_BERT = accuracy_score(y_train_, pred_tweets_train_BERT) \n",
        "accuracy_tweets_test_BERT = accuracy_score(y_test, pred_tweets_test_BERT)      "
      ],
      "metadata": {
        "id": "BiQG7vJ3HS6V"
      },
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "F1_tweets_train_BERT = f1_score(y_train_, pred_tweets_train_BERT, average='weighted') \n",
        "F1_tweets_test_BERT = f1_score(y_test, pred_tweets_test_BERT,average='weighted') "
      ],
      "metadata": {
        "id": "TUGDyhgxLPIJ"
      },
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "d1 = {'metrics': ['Accuracy', 'F1_score'],\n",
        "     'model': ['BERT_by_ktrain','BERT_by_ktrain'],\n",
        "    'train': [accuracy_tweets_train_BERT,F1_tweets_train_BERT],\n",
        "    'test': [accuracy_tweets_test_BERT,F1_tweets_test_BERT]}\n",
        "df1 = pd.DataFrame(data=d1, columns=['model','metrics','train','test'])\n",
        "print('Metrics on disaster classification')\n",
        "df1"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 129
        },
        "id": "pJPazyWRIz4H",
        "outputId": "f5eb0b7f-3dac-4ce6-d3f0-73cd320cd9e0"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Metrics on disaster classification\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "            model   metrics     train      test\n",
              "0  BERT_by_ktrain  Accuracy  0.853530  0.821405\n",
              "1  BERT_by_ktrain  F1_score  0.853102  0.820450"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e505b580-29cc-4faa-9f52-c80b6827026b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>model</th>\n",
              "      <th>metrics</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>BERT_by_ktrain</td>\n",
              "      <td>Accuracy</td>\n",
              "      <td>0.853530</td>\n",
              "      <td>0.821405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>BERT_by_ktrain</td>\n",
              "      <td>F1_score</td>\n",
              "      <td>0.853102</td>\n",
              "      <td>0.820450</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e505b580-29cc-4faa-9f52-c80b6827026b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-e505b580-29cc-4faa-9f52-c80b6827026b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-e505b580-29cc-4faa-9f52-c80b6827026b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot confusion matrix\n",
        "cm = confusion_matrix(y_train_, pred_tweets_train_BERT)\n",
        "sns.heatmap(cm, annot=True, fmt='d', cmap=plt.cm.Blues, cbar=True)\n",
        "plt.title(\"Confusion Matrix on train\")\n",
        "plt.xlabel('Predicted')\n",
        "plt.ylabel('True label')\n",
        "plt.yticks(rotation=0)\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "UlwOrPX6PuiV",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "0614afbe-cd34-4ab5-eb68-19007c010d26"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot confusion matrix\n",
        "cm = confusion_matrix(y_test, pred_tweets_test_BERT)\n",
        "sns.heatmap(cm, annot=True, fmt='d', cmap=plt.cm.Blues, cbar=True)\n",
        "plt.title(\"Confusion Matrix on test\")\n",
        "plt.xlabel('Predicted')\n",
        "plt.ylabel('True label')\n",
        "plt.yticks(rotation=0)\n",
        "plt.show() "
      ],
      "metadata": {
        "id": "SYQ9H_Kp-Xyf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "b2781c26-aa38-4b01-8dd6-c8144e81dc1e"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Inference on BERT Modelling"
      ],
      "metadata": {
        "id": "HrgPT04EJpIt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "pd.set_option('display.max_colwidth', None)"
      ],
      "metadata": {
        "id": "B9-Kpm947xmc"
      },
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_BERT.predict(\"The possible new jerseys for the Avalanche next year. ????\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "id": "_x1MhXHm714-",
        "outputId": "d396541a-82d5-4078-8828-6990a4179896"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 41ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'not_disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[542,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CfR67ICb5Sd2",
        "outputId": "61086ce3-aba0-457a-e63d-32619057fe45"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              The possible new jerseys for the Avalanche next year. ???? http://t.co/nruzhR5XQu\n",
              "classification                                                                         not_disaster\n",
              "Name: 542, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_BERT.predict('Car engulfed in flames backs up traffic at Parleys Summit ?')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "id": "KBxq4Mlf8Xjz",
        "outputId": "c4e66fdf-b7c5-46bf-c79e-20e57255c474"
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 33ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[3264,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5sqTFlFo8OuD",
        "outputId": "b1cc7fb6-ac2a-4ac4-964c-5f26ad09a7dc"
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              Car engulfed in flames backs up traffic at ParleyÛªs Summit http://t.co/RmucfjCaZr\n",
              "classification                                                                               disaster\n",
              "Name: 3264, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_BERT.predict(\"Small casualty on the way to Colorado\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "id": "7wSuv1bHMOwz",
        "outputId": "f5028651-e526-49ec-ac2e-b9900b49239c"
      },
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 34ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[1459,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3VvbOlFn9Jf5",
        "outputId": "4101e7e3-6c79-468b-a724-e1809ec205b9"
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              Small casualty on the way to Colorado http://t.co/hDVmhSQXHm\n",
              "classification                                                        disaster\n",
              "Name: 1459, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_BERT.predict(\"Bicycle-SUV accident in Mesick: Police say that the bicyclist entered onto West M-115 and collided with a boat...\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "id": "WywlvfaJMYs-",
        "outputId": "bd2816f1-b4f5-4aef-a9c4-cbc8c384293f"
      },
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 31ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[1730,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6wY2DbN-9egs",
        "outputId": "ee96ab6c-0641-474c-de77-c2f7316610e9"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              Bicycle-SUV accident in Mesick: Police say that the bicyclist entered onto West M-115 and collided with a boat... http://t.co/A9gtOPyZK8\n",
              "classification                                                                                                                                    disaster\n",
              "Name: 1730, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_BERT.predict('Oil prices falling but drivers could reap benefits')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "id": "ISNvRQi7McZq",
        "outputId": "6dd97864-3c4f-4d25-e37c-18de93e7d56f"
      },
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 35ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'not_disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[1903,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JnwMHsYH91qH",
        "outputId": "bf7c2858-54f3-4ae3-9b67-0f4782de2f4e"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              Oil prices falling but drivers could reap benefits http://t.co/QlTwhoJqYA\n",
              "classification                                                                 not_disaster\n",
              "Name: 1903, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Modelling with Logistic Regression\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "wV9HRFDbLBsv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "(x_tr, y_tr), (x_vl, y_vl), preproc = text.texts_from_df(train_df = X_train_, \n",
        "                                                                   text_column = 'text',\n",
        "                                                                   label_columns=['classification'],\n",
        "                                                                   maxlen=100, \n",
        "                                                                   max_features=100000,\n",
        "                                                                   preprocess_mode='standard',\n",
        "                                                                   val_pct=0.2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nD-6LMeCLR3Y",
        "outputId": "436e1cac-6f52-4e95-cdb5-d3712c8a4b01"
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['disaster', 'not_disaster']\n",
            "      disaster  not_disaster\n",
            "1627       1.0           0.0\n",
            "716        0.0           1.0\n",
            "1171       1.0           0.0\n",
            "7535       1.0           0.0\n",
            "7249       0.0           1.0\n",
            "['disaster', 'not_disaster']\n",
            "      disaster  not_disaster\n",
            "4951       0.0           1.0\n",
            "1386       0.0           1.0\n",
            "7583       1.0           0.0\n",
            "251        1.0           0.0\n",
            "4008       1.0           0.0\n",
            "language: en\n",
            "Word Counts: 16585\n",
            "Nrows: 4872\n",
            "4872 train sequences\n",
            "train sequence lengths:\n",
            "\tmean : 17\n",
            "\t95percentile : 25\n",
            "\t99percentile : 28\n",
            "x_train shape: (4872,100)\n",
            "y_train shape: (4872, 2)\n",
            "Is Multi-Label? False\n",
            "1218 test sequences\n",
            "test sequence lengths:\n",
            "\tmean : 14\n",
            "\t95percentile : 23\n",
            "\t99percentile : 26\n",
            "x_test shape: (1218,100)\n",
            "y_test shape: (1218, 2)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_LR = text.text_classifier(name='logreg', train_data = (x_tr, y_tr) , preproc=preproc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8e4abd5b-98cd-4924-da83-7a47c9efae5f",
        "id": "3TD5tX3jLBsv"
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Is Multi-Label? False\n",
            "compiling word ID features...\n",
            "maxlen is 100\n",
            "done.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "learner_LR = ktrain.get_learner(model = model_LR, \n",
        "                             train_data=(x_tr, y_tr), \n",
        "                             val_data=(x_vl, y_vl))\n"
      ],
      "metadata": {
        "id": "9U5rkWNuLBsw"
      },
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "learner_LR.fit(lr=2e-5, n_cycles=1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "49de40f2-c9ec-4a02-e1de-dac8c63dfeb3",
        "id": "GUdVy_Z4LBsw"
      },
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "153/153 [==============================] - 1s 5ms/step - loss: 0.6906 - accuracy: 0.5757 - val_loss: 0.6888 - val_accuracy: 0.6018\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fb76a89c850>"
            ]
          },
          "metadata": {},
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_LR = ktrain.get_predictor(learner_LR.model, preproc)"
      ],
      "metadata": {
        "id": "LCM4WrU8LBsx"
      },
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "learner_LR.print_layers()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4f8d9de7-ca54-4dcf-9d61-7a16e377c949",
        "id": "qFN1vA9SLBsx"
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0 (trainable=True) : <keras.engine.input_layer.InputLayer object at 0x7fb76a0fbd30>\n",
            "1 (trainable=True) : <keras.layers.core.embedding.Embedding object at 0x7fb76a0fb9a0>\n",
            "2 (trainable=False) : <keras.layers.core.embedding.Embedding object at 0x7fb76a0fb940>\n",
            "3 (trainable=True) : <keras.layers.merging.dot.Dot object at 0x7fb76a0fbc10>\n",
            "4 (trainable=True) : <keras.layers.reshaping.flatten.Flatten object at 0x7fb76a1616a0>\n",
            "5 (trainable=True) : <keras.layers.core.activation.Activation object at 0x7fb76a161730>\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pred_tweets_train_LR = predictor_LR.predict(X_train_.text.values)\n",
        "pred_tweets_train_LR = pd.DataFrame(data=pred_tweets_train_LR, columns=['predicted_label_train'])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5c1b7f9d-ced2-4027-e4ff-a4b3b3acc52d",
        "id": "tNCW0hHMLBsx"
      },
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "191/191 [==============================] - 0s 1ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pred_tweets_test_LR = predictor_LR.predict(X_test.text.values)\n",
        "pred_tweets_test_LR = pd.DataFrame(data=pred_tweets_test_LR, columns=['predicted_label_test'])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "97708aea-c09c-4532-c8a5-dcb457d059d1",
        "id": "XV2DJqZ9LBsy"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "48/48 [==============================] - 0s 1ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "accuracy_tweets_train_LR = accuracy_score(y_train_, pred_tweets_train_LR) \n",
        "accuracy_tweets_test_LR = accuracy_score(y_test, pred_tweets_test_LR)      "
      ],
      "metadata": {
        "id": "I_7i3oInLBsy"
      },
      "execution_count": 55,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "F1_tweets_train_LR = f1_score(y_train_, pred_tweets_train_LR, average='weighted') \n",
        "F1_tweets_test_LR = f1_score(y_test, pred_tweets_test_LR,average='weighted') "
      ],
      "metadata": {
        "id": "HoKm2M3QLBsy"
      },
      "execution_count": 56,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "d2 = {'metrics': ['Accuracy', 'F1_score'],\n",
        "     'model': ['LR_by_ktrain','LR_by_ktrain'],\n",
        "    'train': [accuracy_tweets_train_LR,F1_tweets_train_LR],\n",
        "    'test': [accuracy_tweets_test_LR,F1_tweets_test_LR]}\n",
        "df2 = pd.DataFrame(data=d2, columns=['model','metrics','train','test'])\n",
        "print('Metrics on disaster classification')\n",
        "df2"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 129
        },
        "outputId": "1774212e-70f2-4574-ab2c-c091afbddaea",
        "id": "WEhSzDrQLBsz"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Metrics on disaster classification\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          model   metrics     train      test\n",
              "0  LR_by_ktrain  Accuracy  0.628900  0.586999\n",
              "1  LR_by_ktrain  F1_score  0.629975  0.586999"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-cb4f758f-2ab8-45f2-8cc6-ae8cfd93a9db\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>model</th>\n",
              "      <th>metrics</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>LR_by_ktrain</td>\n",
              "      <td>Accuracy</td>\n",
              "      <td>0.628900</td>\n",
              "      <td>0.586999</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>LR_by_ktrain</td>\n",
              "      <td>F1_score</td>\n",
              "      <td>0.629975</td>\n",
              "      <td>0.586999</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-cb4f758f-2ab8-45f2-8cc6-ae8cfd93a9db')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-cb4f758f-2ab8-45f2-8cc6-ae8cfd93a9db button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-cb4f758f-2ab8-45f2-8cc6-ae8cfd93a9db');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot confusion matrix\n",
        "cm = confusion_matrix(y_train_, pred_tweets_train_LR)\n",
        "sns.heatmap(cm, annot=True, fmt='d', cmap=plt.cm.Blues, cbar=True)\n",
        "plt.title(\"Confusion Matrix on train\")\n",
        "plt.xlabel('Predicted')\n",
        "plt.ylabel('True label')\n",
        "plt.yticks(rotation=0)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "0f74c822-6004-4012-8556-691ee74186f0",
        "id": "q1SCa-dPLBsz"
      },
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot confusion matrix\n",
        "cm = confusion_matrix(y_test, pred_tweets_test_LR)\n",
        "sns.heatmap(cm, annot=True, fmt='d', cmap=plt.cm.Blues, cbar=True)\n",
        "plt.title(\"Confusion Matrix on test\")\n",
        "plt.xlabel('Predicted')\n",
        "plt.ylabel('True label')\n",
        "plt.yticks(rotation=0)\n",
        "plt.show() "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "f51c8595-5730-4f80-bda5-c57ef641c76d",
        "id": "ylwpnzl1LBsz"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Inference on Logistic Regression Modelling"
      ],
      "metadata": {
        "id": "PdVi9IKrOgjg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "pd.set_option('display.max_colwidth', None)"
      ],
      "metadata": {
        "id": "8Wpb9OE-Ogjg"
      },
      "execution_count": 60,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_LR.predict(\"The possible new jerseys for the Avalanche next year. ????\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "outputId": "29dcdaf1-0799-4e9c-c967-80c01f503538",
        "id": "4D5_XYsKOgjh"
      },
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 21ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'not_disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[542,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0fdf1773-59c5-4775-8833-85278c20cd88",
        "id": "kC84amB2Ogjh"
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              The possible new jerseys for the Avalanche next year. ???? http://t.co/nruzhR5XQu\n",
              "classification                                                                         not_disaster\n",
              "Name: 542, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_LR.predict('Car engulfed in flames backs up traffic at Parleys Summit ?')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "outputId": "b47b5ac3-bf5e-4ee7-d3e5-3d6c05e723db",
        "id": "R7J-unAIOgji"
      },
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 19ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'not_disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[3264,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2f81354f-db8b-4bcc-ac9f-1f771bc3093c",
        "id": "iRfDnIzFOgji"
      },
      "execution_count": 64,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              Car engulfed in flames backs up traffic at ParleyÛªs Summit http://t.co/RmucfjCaZr\n",
              "classification                                                                               disaster\n",
              "Name: 3264, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 64
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_LR.predict(\"Small casualty on the way to Colorado\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "outputId": "2e6d0fc7-1038-400e-8578-8ff2aeb06cfb",
        "id": "tfr56FkrOgjj"
      },
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 18ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 65
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[1459,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "477ecba4-b0ec-48e6-a13d-b460c1c7881a",
        "id": "H62tTRlWOgjj"
      },
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              Small casualty on the way to Colorado http://t.co/hDVmhSQXHm\n",
              "classification                                                        disaster\n",
              "Name: 1459, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 66
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_LR.predict(\"Bicycle-SUV accident in Mesick: Police say that the bicyclist entered onto West M-115 and collided with a boat...\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "outputId": "b5b884a7-9618-4c14-a7fc-fb014f53c7f7",
        "id": "sDuyTa9MOgjj"
      },
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 16ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'not_disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[1730,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "61e8e9de-3f36-475e-f3be-3bd43498ae49",
        "id": "x99hAEH-Ogjk"
      },
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              Bicycle-SUV accident in Mesick: Police say that the bicyclist entered onto West M-115 and collided with a boat... http://t.co/A9gtOPyZK8\n",
              "classification                                                                                                                                    disaster\n",
              "Name: 1730, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictor_LR.predict('Oil prices falling but drivers could reap benefits')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "outputId": "6b2eac26-87ac-47cb-c992-ea05bfd7a649",
        "id": "HfBpYCqjOgjk"
      },
      "execution_count": 69,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 17ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'disaster'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_test.loc[1903,:]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "046384b3-3f56-468b-d8a2-53b106ea5fb2",
        "id": "O8rTrpLiOgjk"
      },
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "text              Oil prices falling but drivers could reap benefits http://t.co/QlTwhoJqYA\n",
              "classification                                                                 not_disaster\n",
              "Name: 1903, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 70
        }
      ]
    }
  ]
}